发现自适应任务依赖以实现高效多任务表征压缩
Discovering Adaptive Task Dependencies for Efficient Multi-Task Representation Compression
对做多任务视觉表征学习的人，这篇给出了用可学习相关矩阵推断任务间可预测性、再按动态 DAG 条件编码来去除任务间冗余的思路，可迁移到多任务特征存储与传输、以及多任务模型的特征共享结构设计上。
原文依据 models per-image task relationships and encodes representations following an adaptive directed acyclic graph (DAG)
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论传统图像压缩以像素保真度为优先，但往往保留与下游视觉任务无关的细节。转而压缩任务特定的表征能更好地与任务语义对齐，然而在相关任务之间仍存在冗余信息。现有多任务压缩方法通常依赖静态依赖结构，导致相关任务之间出现冗余的比特分配以及次优的率失真性能。我们提出自适应任务依赖压缩（ATDC），该框架对每张图像的任务关系进行建模，并按照自适应有向无环图（DAG）编码表征。ATDC 通过一个学习到的相关矩阵推断任务两两之间的可预测性，构建动态 DAG 以确定最优压缩顺序，并在其前驱任务条件下对每个任务进行编码，从而实现预测性冗余去除与任务间非对称信息共享。在 Taskonomy 数据集上的实验表明，相较于面向人类的编解码器和最优的多任务压缩方法，该方法在率失真效率与任务准确率上均取得一致提升。学习到的 DAG 揭示了可解释的、依赖内容的任务层级结构，确立了自适应依赖建模作为多任务表征压缩的一种有原则的范式。
英文原文摘要 1322 字符
Traditional image compression prioritizes pixel fidelity but often preserves details irrelevant to downstream vision tasks. Compressing task-specific representations instead better aligns with task semantics, yet redundant information persists across correlated tasks. Existing multi-task compression methods typically rely on static dependency structures, leading to redundant bit allocation across correlated tasks and suboptimal rate-distortion performance. We present Adaptive Task Dependency Compression (ATDC), a framework that models per-image task relationships and encodes representations following an adaptive directed acyclic graph (DAG). ATDC infers pairwise task predictability via a learned correlation matrix, constructs a dynamic DAG to determine the optimal compression order, and encodes each task conditionally on its predecessors, achieving predictive redundancy removal and asymmetric information sharing across tasks. Experiments on the Taskonomy dataset demonstrate consistent gains in rate-distortion efficiency and task accuracy over both human-oriented codecs and state-of-the-art multi-task compression methods.The learned DAGs reveal interpretable, content-dependent task hierarchies, establishing adaptive dependency modeling as a principled paradigm for multi-task representation compression.